Collaborative Spatiotemporal Feature Learning for Video Action Recognition
Open paper
Benchmark
Model
Rank
Results
action-classification-on-moments-in-time
CoST (ResNet-101, 32 frames)
#15
Top 1 Accuracy: 32.4%
Top 5 Accuracy: 60.0%
action-classification-on-kinetics-400
CoST ResNet-101 (ImageNet pretrain)
#132
Acc@1: 77.5
Rank counts only results with a code link.